In some animation programs, attempts to use constraints between objects have caused unintended effects. In certain situations, making or breaking a constraint has caused the constrained object to fly away to an unexpected place in the virtual space of the animation.
For example, in an animation program where each defined constraint would remain in effect (i.e., active) throughout the animation, a rather elaborate work-around approach has been used. Two identical versions of the to-be-constrained object were created: one version that had the constraint and another version that did not have the constraint. The animator would then make only one of the two object versions visible in the animation at any given time, with the other version being hidden. In this way, the animator could deal with the situation that the constraint was constantly in effect.
However, the presence or absence of constraints was not conspicuously indicated in such programs. For example, when the animator adjusted the position of an object in the animation, no knowledge that another object was (at least temporarily) dependent on the moved object was provided or used.